Table 3.

Strategic voting

	Intermediate value
	Low (u m = 3)	High (u m = 8)
Constant (coefficient)	−2.677**	−1.947**
Period/10	0.006	0.003
Simple majority	0.013	−0.035*
Compromiser	0.148	0.306**
Opposer	0.079	0.207**
Rank 2nd	0.207**	0.247**
Rank 3rd	0.542**	0.567**
Rank 2nd × opposer	0.002	−0.056
Rank 3rd × opposer	0.059	0.019
Test rank 2nd = Rank 3rd	χ 2 = 53.5 (p < 0.001)**	χ 2 = 48.8 (p < 0.001)**
Test compromiser = opposer	χ 2 = 0.34 (p = 0.559)	χ 2 = 0.74 (p = 0.390)
Test rank 2nd: compromiser = opposer	χ 2 = 8.53 (p = 0.004)**	χ 2 = 32.9 (p < 0.001)**
Test rank 3rd: compromiser = opposer	χ 2 = 0.220 (p = 0.641)	χ 2 = 1.78 (p = 0.182)

The table presents the results of a random effects probit regression model where the dependent variable is a dummy indicating whether or not voter i in electorate j voted strategically in election t. Formally, it gives the marginal effects derived from the regression model ${Pr}_t^{ij}=\mathit{\Phi }\left(X_t^{ij}{}^{\prime }\mathit{\beta }+{\mathit{\mu }}^j\right)$ where ${Pr}_t^{ij}$ gives the probability that i of j votes strategically in t. $\mathit{\Phi }$ denotes the cumulative normal distribution and X is the vector of independent variables described in the first column of the table. µ ^j is a (white noise) electorate-specific error that corrects for the dependencies across individual decision in the same group. The independent variable ‘Simple Majority’ is a dummy variable indicating situations where one of the preference orderings had an absolute majority of at least 7. The independent variables with an ‘x’ between variables indicate interaction terms. To avoid the dummy trap, the variable indicating Rank 1st voters (i.e., Supporters) has been left out of the regression. The tests depicted in the last two rows test equality of the estimated coefficients. Our results are not sensitive to the choice of quadrature points; when varying these points all differences are smaller than 10⁻⁸

* (**) denotes statistical significance at the 5 % (1 %)-level